EN FR
EN FR


Section: New Results

Data-Driven Systems

Incremental process discovery using Petri-net synthesis.

Participants : Éric Badouel

In [16], we present an incremental process discovery using Petri-net synthesis. Process discovery aims at constructing a model from a set of observations given by execution traces (a log). Petri nets are a preferred target model in that they produce a compact description of the system by exhibiting its concurrency. This article presents a process-discovery algorithm using Petri-net synthesis, based on the notion of region introduced by A. Ehrenfeucht and G. Rozenberg, and using techniques from linear algebra. The algorithm proceeds in three successive phases which make it possible to find a compromise between the ability to infer behaviours of the system from the set of observations while ensuring a parsimonious model, in terms of fitness, precision and simplicity. All used algorithms are incremental which means that one can modify the produced model when new observations are reported without reconstructing the model from scratch.

An artifact model with imprecision and uncertainty

Participants : Éric Badouel, Loïc Hélouët

In the context of the HeadWork ANR project, we started investigating how complex workflows can be defined to handle uncertainty, and use joint knowledge of pools of user to build correct information. The solution proposed so far is a variant of business artifact managing fuzzy datasets. As there are several ways to reach an acceptable final and sufficiently precise dataset, we started investigating equivalence of complex workflows with partial information to allow refinement, enhance performance of data collection, with mastered precision loss.